Myths of Vaccine Manufacturing
blogs.sciencemag.org
blogs.sciencemag.org
Imagine if we had approached this like we did WWII.
We would have nationalized the effort to produce vaccines, building entire production chains and staffing them with anyone in need of work.
We would have built infrastructure to produce n95 masks and shipped them to every address.
What would this mean? We would have saved hundreds of thousands of lives, employed millions of people, strengthened our economy, and would now be wrapping up vaccinating everyone.
Our response to the crisis should be noted as a failure of monumental proportions. The lack of leadership at all levels of government continues to be astounding.
And vaccine production ABSOLUTELY could've been parallelized with approval. In fact, it was. They were stockpiling doses before approval. But the effort was at least an order of magnitude smaller than it ought to have been.
We constantly make excuses for this lack of effort. We shouldn't. We know what is possible. We can read history books.
That's incorrect.
4 months before the Pearl Harbour, FDR instated the Supply Priorities and Allocation Board by executive order. The purpose of the Supply Priorities and Allocation Board was to coordinate the distribution of materials and commodities related to national defense and to assist the Office of Production Management (OPM) in carrying out their overlapping duties.
On january 1942, merely 4 weeks after Pearl Harbour, FDR dissolved the SPAB and instated the War Production Board by executive order.
The WPB directed conversion of industries from peacetime work to war needs, allocated scarce materials, established priorities in the distribution of materials and services, and prohibited nonessential production. It rationed such commodities as gasoline, heating oil, metals, rubber, paper and plastics.
Moreover, the U.S. economy was already radically shifting before 1941. Between 1939 and 1941, unemployment feel from 17.2% to 9.9% and by 1942 it feel to 4.2%.
There's also an implicit assumption in your statement: the U.S. not actively joining the War until june 1944. That's incorrect as well.
The U.S. already actively engaged in the North African Campaign on 11 May 1942. Barely 6 months after Pearl Harbour.
Pearl Harbour kicked off the Pacific Theatre right off the bat. In June of 1942 - 7 months after Pearl - the Japanese Navy was soundly defeated in the Battle of Midway by Admiral Nimitz. August 1942 - 8 months after Pearl - the U.S. took the offensive at Guadalcanal.
Admiral Yamamoto, who led the Pearl attack, never believed Japan would be able to win from the United States in the first place. He believed it was a massive strategical mistake. He has conveyed his feelings:
> "I can run wild for six months ... after that, I have no expectation of success".
Which is exactly what happened.
So, no, the involvement of the United States in World War II is anything but an example of a nation state being forced to wait until it has ramped it's economic output before it can act.
https://en.wikipedia.org/wiki/War_Production_Board https://en.wikipedia.org/wiki/United_States_home_front_durin... https://en.wikipedia.org/wiki/North_African_campaign https://en.wikipedia.org/wiki/Battle_of_Midway https://en.wikipedia.org/wiki/Guadalcanal_campaign https://en.wikipedia.org/wiki/Isoroku_Yamamoto%27s_sleeping_...
But that's not what happened in WW2. Businesses were not nationalized and they weren't staffed by anyone in need of work.
The War Productions board simply directed companies on what to produce. They could run their business as they see fit with some controls like wage caps, raw material quota, production quotas.
And it works better for war - switching from making planes to airplanes versus having companies switch from making microchips to making vaccines.
...well, because it's NOT open. The very bottleneck he's talking about is something he has to wave his hands about because it's bespoke, proprietary, and not well-published or open. This inadvertently validates the original tweet!
https://en.wikipedia.org/wiki/Government_patent_use_(United_...
Additionally, until this brand new mRNA vaccine technology actually demonstrated efficacy it was not quite so clear cut that we should put a huge proportion of our resources into making more of these devices, rather than explore different vaccine platforms.
But we're not really limited in resources, here. We have TONS of money, people are unemployed, and this is essentially the biggest problem facing the world right now. We can AND OUGHT TO be "wasting" money by throwing large resources at a wide range of potential vaccines (some of which wouldn't work).
The whole "we don't have resources" thing is absurd. We've been spending literally trillions of dollars on COVID-19 relief (and monetary easing, etc) which is a poor substitute for actually solving the COVID-19 problem.
Furthermore. Since the process of vaccine development and manufacturing is closed, done in the shadows, no one had any idea where the bottleneck will be. (Yes, some companies bet on their vaccine and started expansion while the development was ongoing. And Derek speculates that the supply chain is at full capacity, but ... we don't know, and again, we could have been expanding that too exactly since a year ago.)
The mRNA LPN packaging might be new, but manufacturing plasmids that churn out Adeno Associated Virus particles with the correct payload is so old tech random youtube dude did it in 2018 February.
For example Trump simply went with ordering only 100 million doses of the Pfizer vaccine and said meh when asked about whether maybe in a country of 300+ million people they might happen to ... need a few extra ones.
The EU negotiated so hard that even though it ordered 400M from AstraZeneca, somehow they find it more important to fulfill other orders first.
It's too little too late. I understand that Bill Gates [0] can be so blasé easily, as for him it's just more of the same. Plus he has predicted it, he was directly trying to do something against these global health problems for at least a decade now. But this "inefficiency" - that thousands of people die of each day - is mind boggingly infuriating.
I can't believe this is true.
The Pfizer and Modern vaccines were clearly in the top 10 last March and April. How could you argue that given 10 months and a billion dollars each we couldn't double our microfluidic device capacity for these two vaccines.
Also, we did double our microfluidic device capacity. We more than doubled it. We increased our capacity by something like a million fold. Before these vaccines (first of their kind), there was no commercial scale capacity. This has all been invented in the last year. There was even a chance that achieving such capacity would not be possible. It is fairly common that drug candidates that look promising in phase 1 or 2 simply cannot be manufactured at scale with the quality required to ensure consistent safety and efficacy.
[0] https://www.ncbi.nlm.nih.gov/pmc/articles/PMC7583697/figure/...
[1]https://en.wikipedia.org/wiki/Operation_Warp_Speed#Companies...
We did have it, we just didn't know which one it was. And we refused to risk any individual life in order to potentially save hundreds of thousands or more.
We also could have invested single digit billions early on to build capacity for all of these different potential vaccines, but we decided to play if "safe" and will now be spending over a trillion again to try to save the economy.
I don't blame the pharma companies for this. Our government and medical establishment was not intellectually prepared to make the hard decisions required to save us. And we need to be building momentum to learn how to do better.
Are we sure "we" didn't? I don't know the actual numbers behind what was spent ahead of time for the various vaccines, but I believe the manufacturing process was ramped up before the trials completed.
It's possible that I'm wrong! Maybe we invested everything possible and couldn't have done significantly more. But it's striking that this article doesn't even address that question.
Basically, I haven't seen any evidence that this take is wrong: https://thezvi.wordpress.com/2021/01/28/covid-bill-gates-and...
While initial production ramp up would get the richer countries vaccinated quickly it seems like Gates has a more holistic picture and wants to see the whole world vaccinated faster.
I agree with this thinking. It won't matter if rich countries are vaccinated if other countries are coming up with new variations all the time. We want to eradicate this by having everyone in the world on the same page, not have it hanging around like the flu.
https://bfi.uchicago.edu/wp-content/uploads/2021/01/BFI_WP_2...
There was limited capacity to manufacture back then, probably less so than there is today. Which vaccine candidate would Pfizer have manufactured? All of them?
And then, yes, a lot of those factories would have ended up sold for scrap (or mothballed for future pandemics or mutations). But the few that worked then save trillions of dollars of value and hundreds of thousands of lives.
It's a simple matter of calculating expected values and investing accordingly. But thats not how our civilization works.
The Pfizer vaccine seems to have a shelf life of 6 months, so realistically the earliest they could have begun mass manufacture and have an effect today would have been June. That's right around the time they narrowed down to a single vaccine candidate.
I'm surprised that mass manufacturing didn't begin back then with the only possible candidate, especially since the US government also put their order in around that time. I can't find a whole lot of info on when they did ramp up (something I saw said October), and what reasons they had for not starting earlier.
But we are risking individual lives now. Vaccines have side effects. It's never about an individual life, always a calculus of number of lives risked vs number of lives saved, I would think.
The cost of our approach is that roughly the same proportion of the country's population must bet infected as in the trials.
Back in my time at uni, volunteers had to be unpaid to ensure they do it kf their own free will. Its unclear which is better / worse.
It was clear the US government made the wrong decision when they backtracked and tried to buy more doses from Pfizer + Moderna[1]. Given how much the US has spent on stimulus, you would think even like 10-15 extra billion would be nothing.
[1]: https://www.nytimes.com/2020/12/07/us/trump-covid-vaccine-pf...
Are you volunteering? Will you sign all the paperwork?
If you die in the trial or get a severe reaction, would your family be happy with the paperwork, or they will claim that they fooled you?
If you die and the vaccine candidate fails, will the TV claim that they should have used a good old method like a modified adenovirus instead of playing god and creating a frankenvirus in the lab?
Even in this pandemic, we've (correctly, I would say) basically thrown out important principles like freedom of movement and freedom of association during lockdowns. I think that allowing volunteers to put themselves at a relatively low risk to speed up vaccine trials would have been a relatively small transgression, compared to the lockdowns and travel bans (which, again, I think were justified).
But I do take your point.
Yet, things like enforced mask wearing, or enforced out of home quarantine/isolation somehow were off the table. It is a weird path-dependent quirk of the fact that we had no testing at the beginning, that instead of quarantining and isolated the exposed and infected, we are effectively quarantining everyone.
It's just that you have to wait a few months to know if the vaccine you're testing is effective. You can't make this delay shorter with more volunteers.
And when you're fighting a decease that "only" kills 0.5% of the people inflected, your risk margin is pretty low (what if your vaccine creates deadly consequences to 0.6% of the people vaccinated?).
The mRNA vaccines had to wait until November to get enough infections.
There can be an issue that the measurement of the effectiveness of the vaccine is then related to the exposure protocol (which may not be the same as typical natural infections), but it's reasonable to expect results much sooner.
Challenge trials need far fewer people, so paying each one $1M is feasible.
The challenge arm wouldn't have any influence over the other arm (and likely not much impact on recruitment), but might provide results for some groups much faster. Starting vaccinations on younger healthcare workers in August seems like it would have been a win (assuming they had data to justify it by that point).
Depending on where in society you sit probably shapes whether you think it was worth it. (Full disclosure, I do but I was also not inconvenienced much.)
As it stands currently, I note that interventions in Australia are much faster now.
Our stand down position was we had to wear masks to the supermarket and on public transport. Everywhere else was pretty close to normal.
But one case detected yesterday in our 4 million population city, means we all wearing masks whenever we're indoors today.
That one case has flicked that switch overnight. So it seems that the main thing is to test well and react quickly.
If they'd acted early and decisively like the countries you mentioned, the outcome would likely have been better. But it would still look very different from those countries.
Arguably the West handled in the worst of both outcomes - arbitrary and incomplete lockdowns - that caused a huge financial toll and a huge human toll.
We do pay a price for those freedoms sometimes in times of crises - but there's no doubt which system we'd rather live under.
They had an extra month of warning compared to Europe, that made a big difference.
The USA had the poorest performance. When a country can't attend to basic matters of engaging with reality and protecting its citizens, despite incredible resources, it is defunct.
I am going to rant here. This virus is really an indictment on the state of the world. It seems inevitable now that it will be endemic, and due to the infectivity and presence of animal hosts, will not be eradicable. This is terribly disappointing. Consider that this is the year we gifted an entirely new disease to all the billions of people that will live in the future. We just increased by 1 the number of things that can make you sick and kill you. Think about HIV - if we could go back and stop HIV from spreading, with the benefit of 40 years of hindsight, we absolutely would have done that, no question, not for one minute. The economic saving alone is compellingly enormous, not accounting for the amount of suffering that would have been avoided. To me, all this calculus about economic activity and mental health versus stopping the virus is so short-sighted and missing the point entirely. It seems obvious to me that any degree of short term pain is a mere blip compared to how many people will suffer and die of the virus over the next 20 years because we let it out. We had one chance, one moment to stop this virus and we blew it. Everyone, every country, ever politician, blew it. I know it could have emerged in a few years, or there could be another virus next year but that doesn't change the fact that we could have stopped this one.
What we had was the ending of Indiana Jones and the Last Crusade.
A table full of cups. Some of those have the Water of Life. Some just water. Some will outright kill you. Pick one.
It's only when you go through the trial process that you can separate the life-giving stuff from the rest.
As of the end of October, the U.S. had invested $18 billion to "build capacity," mainly by making advance purchases of vaccines that had yet to be produced (and had yet to pass clinical trials). See [1]. But maybe you're thinking of building capacity in a different way.
> We refused to risk any individual life in order to potentially save hundreds of thousands or more.
This is such an important point. In particular, human challenge trials, in which healthy, low-risk volunteers are infected with a low dose of the virus, could have saved tens of thousands of lives -- at a minimum -- in the U.S. alone. The case for HCTs, including the ethical case, is taken up in [2].
In a comment below, someone likens HCTs to the infamous Tuskegee syphilis study. The comparison isn't apt. In the Tuskegee study, people infected with syphilis were promised medical care that they were later denied. When they were diagnosed with syphilis, they weren't even told of the diagnosis. Those conditions are a world apart from the HCTs that Eyal et al. propose in [2]. - - - - -
[1] https://www.bloomberg.com/news/features/2020-10-29/inside-op...
[2] https://academic.oup.com/jid/article/221/11/1752/5814216
I.e. the way we do things means we cannot stop covid.
And then on the other side of it, Moderna has a vaccine against the South African mutation in a trial. And production capacity on the mRNA vaccines is pretty clearly not maxed out in any long term sense (and then the other vaccines apparently have a better path to volume; J&J still sounds like they will make 1 billion vaccines this year. The viral vector vaccines aren't as immediately adjustable as the mRNA vaccines, but they are similarly targeted, choosing a protein to be expressed).
In any case, the more people who are vaccinated, the more likely a strain will arise that is not affected by the vaccine. This will reset all the vaccination efforts back to zero.
This is what I am referring to.
I guess there's lots of room for arguing about definitions, but I think we don't go back to zero if there is a vaccine escape. The production capacity that has been built in the last year is a big step forward (and we can expand it), and we are building out the administrative capacity to get the vaccine in people's arms (capacity that won't just vanish).
The concern wasn't Trolley problem[1] paralysis, where authorities are afraid to deliberately shift harm to a minority to protect a majority. The concern was vaccine-associated disease enhancement (VADE), which could possibly cause more overall harm and deaths than simply doing nothing at all. It's a significant and legitimate concern because not only has it happened before with released vaccines, it has been a particular problem in SARS vaccine research and trials for 20 years. See "Learning from the past: development of safe and effective COVID-19 vaccines", https://www.nature.com/articles/s41579-020-00462-y
We had complete candidate vaccines mere days into the outbreak, but nobody was sure they were safe because they were just the latest iterations in long lines of similar candidate vaccines, all eventually failing. They couldn't rush trials too quickly because VADE situations might not become apparent without a large pool tracked over an extended period of time so that you can see what happens with reinfections, etc.
We got lucky. It seems to be the case that we had just recently turned the corner in resolving many of these barriers. If were were facing COVID-10 (i.e. a SARS pandemic in 2010), we'd be screwed because in 2010 we were much further away from figuring out how to avoid VADE-like failures in SARS vaccines.
The FDA works in a whole other way.
There have been plenty of fumbles and mistakes all around, but I don't understand how people don't appreciate how incredibly fast the process has already been.
IMO, the real issues involve more mundane matters of logistics--coordination, resource allocation, etc. Could the process have gone faster? Probably. Not because of safety protocols or inflexible regulations, but because we had unforced errors starting and executing trials. For example, AFAICT, trials could have started in Brazil in early June, but presumably didn't because the companies needed more time to spin things up. And of course manufacturing and distribution of the vaccine should be going much smoother. Pharmaceutical companies and, especially, governments should have thrown more resources toward growing the supply chain of machinery; it was foreseeable from the beginning because of the manifest deficiencies in the supply chain for personal protection equipment, ventilators, etc, all of which made evident the need to retool industries to be able to scale manufacturing. And governments have had almost an entire year to figure out distribution protocols, but instead it seems officials across the board in the U.S. simply assumed it would be sufficient to push vaccines through existing healthcare networks, despite the fact they're notoriously byzantine and inefficient, especially for something so time critical, and especially given the need for consistency and uniformity across the population.
Is COVID-19 providing cautionary tales regarding government intervention in the free market? Absolutely. Not because governments have been intervening too much, but because they've been intervening too little; not because it has proven government more inept than private enterprise, but because private enterprise is intrinsically incapable of achieving alone the level of coordination needed to scale our response in the time required, yet we have shown ourselves too politically and fiscally risk averse to wield the tools of government in an emergency. It's like the delayed and anemic Katrina response, writ large.
If this is true this is the most insane thing I've ever heard.
They can't have it both ways. You can't say "the FDA knew it was safe and effective so they gave the geeen light to the manufacturers" and also "they had to withhold this vaccine from the population because they didn't know if it was safe and effective".
> Is COVID-19 providing cautionary tales regarding government intervention in the free market?
I don't believe criticism of the government needs to be condensed into "government intervention good" or "government intervention bad".
I don't think the solution is smaller government or bigger government it's better government.
The FDA has royally fucked up over and over again by being far too cautious. A lot of the government fucked up when they said masks didn't work.
The CDC fucked up when they promoted vaccinating essential workers instead of the old.
And the executive branch has been far too cautious and uninvolved in ensuring we have enough high quality masks, and broadening vaccine the pipeline.
It's like saying "We had steel mills; we had cars back in February!" All it takes is designing and building them.
Part of the problem may be that wafer fab people and drug production people don't talk to each other much. Someone should get them talking. Might come to nothing. Might lead to "oh, we can make that easily".
I suspect if you track down what's needed, you'll find that all the manufacturing capability for making more of those machines is already in use by the vaccine makers. Bespoke microfluidics devices aren't common or easy to make, although some components are.
Each new human innovation is based on a massive mountain of previous human innovation. Many, many companies are on the edge of this innovation and allowing them to use the innovations of other companies without fear of patents litigation would allow them to make that last step now.
Patents are one of the most insidious harms we have in our society today. Patent defenders claim that patents protect innovators. The truth is actually the exact opposite: patents are a monopoly legal tactic most often used so that the patent obtainer does not have to innovate further or execute. Patents kill all further innovation in the area they are in. A great example of this is that hundreds of companies used to manufacture aircraft. Boeing and Airbus bought up all the companies that own patents, and now it's nearly impossible to start a new aircraft manufacturing business due to Boeing's monopoly on the patents. Aircraft have changed very little ever since.
We need to eliminate the patent system and allow free trade to actually work.
That's a general rule of thumb, but it's not a law. There are innovations that are non sequels of previous developments. Sometimes inventors think differently than anyone else.
The idea that multiple companies were "almost" on the edge of mRNA vaccines is pretty much proven false by the place the two existing companies got the tech - from the pair of people who developed it, one for each company.
That's billions of people who will be incubating more mutated variants of COVID-19.
And secondly, we'll need that equipment for COVID-19 variant booster shots.
(And will come in handy for vaccinating with a universal flu vaccine, even HIV, etc.)
I was involved in a relatively simple tech transfer of a pharmaceutical agent to a plant in China. We pulled it back to the US because every single batch failed. And it wasn't due to shoddy workmanship since we had US people on the ground directing everything.
What everyone else seems to be saying is that Covid-19 has a bus factor of like 5 or 6 people, who if they died,would lead to any breakdown in the process becoming an irrecoverable incident.
That single point of favor needs to be distributed, not just for production capacity, but for the future safety of the human race.
Call me skeptical. I think it's much more likely that this falls under trade secrets.
This "nationalizing essential drugs" business could set a dangerous precedent after all.
https://www.precisionnanosystems.com/platform-technologies/p...
Here's a company which makes the "chips" which go in such cartridges.[1] And here are some chips.[2] Making those involves etching tiny patterns in glass, lining the channels with various coatings, and assembling the layers of glass into devices. All this has to be done in a very clean environment. Note the similarity to the processes in a wafer fab.
The point here is that the semiconductor industry has many processes and tools which might be helpful if there's a need to manufacture microfluidic components in quantity.
[1] https://www.dolomite-microfluidics.com/
[2] https://www.dolomite-microfluidics.com/product-category/micr...
Which is really the issue with all potential fixes: either they are unknown and thus unacceptably risky for something we will use on hundreds of millions (or billions) of people, or if we take the time to get to acceptable risk it will be too late to really help with this issue.
I knew in February or March that the mRNA vaccines were extremely promising. NIH experts knew earlier than I did.
The mRNA approaches had a head start because they had been studying MERS and SARS-1 before COVID-19. They hit the ground running once the sequence of the virus was released. It was known nearly immediately that they were promising approaches. That there were hyper-conservative uberskeptics that didn’t think so (and hadn’t actually been following such approaches in detail) and snootily spoke in the press about that doesn’t change what the real experts who were keeping abreast to the latest developments thought.
Also the instrumental parameters from manufacturer A probably cannot be applied directly to the instrument from manufacturer B. It is an entire process/system that can not be reproduced overnight. It takes time.
https://www.statnews.com/2020/12/11/covid19-vaccine-timeline...
"“Vaccine manufacturers hold their production capacity pretty close to their vest because it’s a point of a negotiation.” he said. Companies want to have flexibility in their contracts so they can balance production of various drugs and vaccines. “These companies are businesses and want to maximize their revenue. They’ll continue to make other products they can sell while manufacturing their vaccine.”"
I know less than either Lowe or the State News author. I don't know where the particular bottlenecks might be. But it seems hard for me to believe that "ramping up is impossible" doesn't mean "ramping up is impossible under standard procedures, where we keep proprietary information proprietary, where we don't spend massively on capital equipment we might not need down the line (etc etc)"
That is to say that we're far and far from a Manhattan Project intended to push vaccine manufacture to the logistical maximum. That's just how just any institutional process works.
It seems very credible to me that ramping up from (guessing) thousands of doses to billions will take a long time.
As I understand it, once the industry is set up, producing mRNA vaccines should be substantially faster than traditional techniques, since there is no biological step to wait for. They're "printed, not grown".
There is an extremely easy solution to this problem, which is, just paying drug companies so much that it’s not worth for them to make anything else. Unfortunately, the governments around the world have failed the easiest cost-benefit analysis in the history, that is, trillions > billions, and so they negotiated the vaccine prices to be rather low, saving billions, while at the same time, spending trillions in stimulus money.
Imagine if they just allowed the companies to sell the vaccines at the prices dictated by the market. Also imagine if they allowed AstraZeneca vaccine in the US without jumping through extra hoops, which cannot possibly give us more valuable information than we already have, now that millions of doses have already been administered in UK. Alas, the regulatory state has failed us in the pandemic the worst it has ever had.
Instead, the current arrangement, where the governments claim first dibs on hundreds of millions of doses, while paying relative peanuts compared to the insane amounts of value those vaccines could produce, do very little to incentivize the companies to spare no expense to accelerate the production. Once they satisfy the obligations they have with governments, remaining buyers won't want to pay much either, so why bother?
But I bet if this were legal, the vaccine companies would find overnight ways to produce more vaccines. Instead, we have severe shortages. Shortages and gluts always happen when the government tries altruism instead of practicality.
In the counterfactual case of pure market driven vaccine pricing, a couple of election cycles from now there would likely be some politician rising on angry rhetoric against the approach. "Big Pharma made billions in windfall profits on human suffering -- let's claw back the money and vote out their government cronies who enabled it."
I'd like to say that the electorate wouldn't fall for shallow bombast like that, but I can't. People got mad at Gilead for the pricing of their hepatitis C cure even though it had lower lifetime cost than the previous, far less effective state-of-the-art treatment.
The federal government should massively invest in MRNA manufacturing or whatever other vaccine technologies make the most sense and lease the facilities out at a loss if need be to pharma during normal times. Or just subsidize pharma to have massive amounts of capacity ready to churn out 5-10 million doses a day. I don't really care how it happens, I just want the capability as a society and I'm willing to pay for it.
$2 billion for Oxford.
$3 billion for J&J.
$20 billion for Novavax.
$24 billion for Pfizer (at worst price).
$48 billion for Moderna (again, worst price).
You end up with a billion vaccines you don't need, you give them away.
This is likely an over-investment if you consider the US in isolation (for instance, all 5 vaccines work well to reduce symptomatic and serious infections).
Source: https://www.bbc.co.uk/news/health-55274833
https://www.telegraph.co.uk/politics/2021/01/31/britain-give...
It sounds as though you're talking about the advance purchases that the U.S. made as part of its "Operation Warp Speed" -- except that you (reasonably) would have liked those purchases to be even larger than they were.
See especially the table in https://www.bloomberg.com/news/features/2020-10-29/inside-op...
For the less expensive vaccines it doesn't make any sense at all that they ordered 100 million doses instead of 300 million. For the mRNA vaccines I have a lot of questions about how hard they tried to identify potential bottlenecks in say, July (but I don't think there is much information available to decide in either direction.).
I link a working paper in another comment where some economists estimate that we left an awful lot on the table. The US less than the globe, but given the shared benefits, this would be a great issue for a wealthy country to take a leadership role on.
Color me a little skeptical. This looks like a big GMP bioreactor facility. Once the vaccine is made, how does it get into vials? Fill and finish production lines are usually another massive complex. Are they going to transfer it offsite for packaging?
And this facility won't be finished until July 2021. Then the tech transfer has to happen, trial runs completed, everything validated. They might start actual production at the end of 2021.
Tech transfer of pharmaceutical processes is no joke. I've done it before.
I don't know if you've been there, but the facility is pretty massive. I don't think packaging will run into space concerns.
There's actually a vaccine plant right next to it, which is dwarfed by the NRC facility (of course not all of it is dedicated to vaccine production)
Once those are resolved, there's still a fixed amount of time that it takes to convert existing production facilities to some new purpose that you can't really eliminate (the old saying about 9 women having a baby in 1 month applies here...)
Basically, it's the sort of thing HN readers don't like to hear: the complexities of the real world create obstacles that are difficult to surmount with technology alone. If I had to guess, even if you got approval to have N partners manufacture the vaccine today, it would still take a few months to ramp those partners up to capacity.
https://www.biopharma-reporter.com/Article/2020/08/06/AstraZ...
https://www.reuters.com/article/health-coronavirus-canada-va...
Another traditional vaccine, Novavax, is signing licensing deals as well:
https://globalnews.ca/news/7613878/covid-made-in-canada-vacc...
He didn't really answer--can that one be scaled since it has more standard dependencies? If so, isn't the "naive" question perfectly valid in that case?
That said, I'd prefer to get the mRNA vaccines as they seem much more effective, so any scaling up done there is still useful even if J&J alone could get rid of Covid in the US in a few months. (pending need to vaccinate kids for herd immunity, Pfizer is the farthest along that path, but not close to mass vaccination)
I am still unsure which route seems best.
Don’t mRNA vaccines target the spike protein?
The spike is a topic of debate currently as mutations/changes to the spike (South African variant etc) may affect mRNA vaccines more?
Moderna themselves are developing a booster shot targeting the SA variant & a separate booster for all the other variants (U.K. et al) for this very reason (read: you’ll need 2 doses of vaccine & 2 booster shots for full protection against all current forms of COVID/virus)
So the mRNA route might well be 4 shots for just the variants we know about.
That isn’t to say non mRNA vaccines (AstraZeneca, J&J) won’t also have the same issues (decreased effectiveness against the mutated virus ) but from what I’ve read it seems more likely to affect mRNA
You are correct. That being said, the targeting of the spike protein for the adenovirus vaccines is different to mRNA vaccines[0]
I have done some more reading & have seen the recent news of AZ & J&J having less effectiveness against the South African variant. With one positive for J&J being that their trial ran later than the other current crop of vaccines & part of the trial was done in South Africa - so more data is available[1].
Additionally, mRNA vaccines have been associated with instances of anaphylaxis[1], which has yet to be observed with the adenovirus vaccines (which could be due to mechanism, or that they have been used on less of the global populous to date). Furthermore, there is a strange "benign" rash appearing on peoples arms after being administered the mRNA vaccine - we are being told it's harmless, but also that nobody knows what it is precisely. You can look up "COVID arm" for details.
[0] https://www.nytimes.com/interactive/2020/health/johnson-john...
[1] https://www.kqed.org/science/1972627/the-differences-between...
> J&J doesn't really need to be scaled up because they alone have the ability to produce so much of it.
I'd bet 100 dollars that J&J ships less than a billion doses before July.
Are you willing to take the other side of the bet? If I'm right, that means that they could have "used help" even though they "have the ability to produce so much of it."
Edit: Before you take my bet, which I'm happy if you do, you may want to read this:
https://www.nytimes.com/2021/01/13/health/covid-vaccine-john....
In a global pandemic, why are we guessing?
Are the specs for the design of the "Vaccine Machine" machine-- as he terms it-- public knowledge?
The pun in my statement was that Mr. Derek Lowe is writing a popular series of blog postings on dangerous chemicals. It is called "Things I Won't Work With".
https://www.bing.com/search?q=derek+lowe+things+i+won't+work...
Why are we guessing about trade secrets in a pandemic?
Seems to me that an advanced society would have the sense to say that while trade secrets are obviously an important part of the economy, in the case of a global pandemic we can suspend them specifically for the design specs of the bottleneck in vaccine production.
And we should have done large-scale phase 3 trials by randomizing distribution of large fractions of the output of these megafactories.
It's just completely wrong to say that everything was done as well as it could possibly have been and there's no way it could have been faster. There basically no amount of money that it would not have been worth it to spend.
This whole vaccine availability "scandal" is driving me nuts. It was clear that nobody knew which vaccine would work. It was prudent to buy all of the potential candidates.
It was also clear that there would be a ramp up phase, in each country and for each vaccine. The EU pretty much delivered on part one, sourcing. And I don't care what the press writes, 5 doses for each EU citizen regardless of age is enough.
EU member states screwed up on managing part two. And communication. I think, things will tun smoother once we have a vaccine that can be handled and used by every doctors practice. That supply chain works for millions of vaccinations every year. Before that, it is hit and miss.
That mRNA completed phase-3 trials first was a big shock to everyone. That they are significantly better than the conventional vaccines is also a big shock. If we knew about these shocks 1 year ago we would have invested a lot more money in mRNA (and none at all in all the other vaccines - some failed trails, some are in progress, and the rest are are not as good) and had more manufacturing as a result.
You can prepare for everything but it needs to be done on the basis of balancing many different concerns.
Instead, the administration sat on its hands and said "what could we possibly do?" and the mess we have today (in the US at least) is the direct result.
Somehow the investors need to get back their investment, cover their risk, and make a profit on top of that. Impossible?
Even worldwide, the median per-capita household income is $2,920. 1 trillion is a lot of money.
Edit: is there a name for the fallacy “something costs an individual $x, therefore the individual can afford to pay $x”, or the similar fallacy “something costs people an $x GDP, therefore the government should pay $x to fix it”.
I don't know what you are talking about. Speculatively building out capacity for as yet unproven vaccines was suggested and planned back from the beginning already.
(Here for example the Gates foundation in April 2020: https://www.businessinsider.com/bill-gates-factories-7-diffe... )
> There basically no amount of money that it would not have been worth it to spend.
The Economist calculated, based on a Goldman Sachs study, that "an American wearing a mask for a day is helping prevent a fall in gdp of $56.14. Not bad for something that you can buy for about 50 cents apiece." - so, basically a 10000% ROI in a day!
Yes, then, money could have been sensibly employed, but on what? Maybe it would have been better employed on PR to persuade people to wear masks, or maybe a DDOS shutting down social media. How to allocate money optimally in real time remains a difficult question.
I don't think things could have been substantially faster, though things could have been much less painful if not for the intransigence and recklessness of many covidiots.
EDIT to add the source: https://www.economist.com/finance-and-economics/2020/08/22/w...
and outline if paywalled: https://outline.com/yCeVzn
Maybe!
But, while the transmission rate appears to be sort of high - the fatality rate seems to be low (and getting lower as we figure out effective treatment protocols). I guess I personally don't think it was worth killing the economy over it.. Isolate "at risk people" (say over 75-or-so?) In fact, at least in the US - I think the shutdown caused more harm than the disease (although reasonable people often draw different conclusions from the same data...)
With covid, we have a mortality rate of something between 0.3% and 3%³. The CFR of influenza includes - to the best of my knowledge, since i can't find any sources - vaccinated people. So without vaccination, influenza would have a CFR around 0.8% * (100% / 45%) ~= 1.78%. This is within the error bars of covid19 CFR, since there was no vaccine until now.
I don't want to downplay covid19. I do want to question the effect on the society, and the personal responsibility of everyone involved.
@Downvoters: just state why please.
If you disagree, please be so kind and educate me why.
¹https://en.wikipedia.org/wiki/Influenza
The Wikipedia article has a graph, which I assume you're referring to with this figure, which shows the mortality rate of 0.8% only applies to those ages 65+. The blended rate for a typical flu season, given in a table below, is <0.1%.
And CFR, case fatality rate, won't include vaccinated people who didn't get sick so your final calculation doesn't make sense.
What does this mean?
There are many factors at play here. Aside from mortality, you also have to look at the R0. Covid 19 has a significantly higher R0 than seasonal flu (estimated 1.5 vs 2.5+).
In addition, many people have some immune resistance to seasonal flu, but there was no existing resistance to Covid 19.
You only have to look at the health situation in Italy, Portugal, UK and elsewhere to see the difference between the two. Or look at the excess death graphs.
I doubt our response is good. We must protect vulnerable people, and we have the means to. But this is what we are neglecting, and this is what causes deaths.
And while we're not protecting vulnerable people, we're also harming (to a not yet known degree) people who are not vulnerable at all: the young. The're being deprived of their education (at least in europe), of their young years, of living their young life. Even my parents in their 60s say they'd rather protect themselves or die instead of causing depression among the young.
But who cares.
So, with your 0.3%-3% CFR, Covid-19 is three to thirty times as deadly as flu. Call it ten times as deadly; it is also more virulent (R0 of 1.4 vs 3-5 for Covid-19).
That's why it is being treated more seriously than flu.
From what we've seen, yes, the mortality rate of COVID19 is similar to the flu. However, the mortality rate of the flu is without other interventions aside from the flu vaccine. In a regular flu season the public doesn't social distance; they tend to group even closer indoors because of the cold weather. The US has never normalized mask wearing; people in the beginning stages of the flu will go about their lives, and work places actively encourage people to continue to work in-person unless they are too unwell to do so. I would wager most people in the US don't improve their hygiene during a regular flu season either. We've done all that (to an extent) with COVID19 and we've barely managed to make an impact.
Look at some other important data points. Average US flu deaths / year: 36k [1]. COVID19 deaths in the US in only 11 months: 445k. [2] If we assume the first COVID19 death in the US was in early February as some report, that's over 12x the deaths/year than the flu. If current trends hold (~3k deaths/day for the next 27 days) until March 2 (the first officially confirmed COVID19 death in the US), we'd be at 526k deaths, or more than 14x as many deaths/year as the flu. If this ran at this rate for the entirety of the next 365 days, we'd hit nearly 1.1 million deaths, or over 30x as many deaths/year as the flu. That's assuming we don't have any massive improvements with vaccine rollout.
The contagiousness of the two is also orders of magnitude different. Flu is estimated to have an R0 value of ~1.5 [3]. Some estimates put COVID19 at an R0 value of 5.7 [4]. (Remember, R0 is without interventions). The way to estimate "herd immunity" rates is the Critical Vaccination Threshold formula: (1 - (1/R0)) / E, where E is the effectiveness of a vaccine. With an R0 of 1.5 and an efficacy of 67% (from your second link. Remember, this is the effectiveness of the vaccine, not the effectiveness in a real-world population, which factors in vaccine rates), we would need roughly 50% of the population to vaccinated to start preventing illnesses in those that are unvaccinated. With an R0 of 5.7 and an efficacy of 94% (Pfizer vaccine), COVID 19 would require nearly 88% of the population vaccinated to achieve the same rate. Even using the lower-quoted R0 value of 2.5 (R-value _with_ other interventions, such as social distancing, mask wearing, improved hygiene), we'd need nearly 64% of the population vaccinated.
You're also ignoring the fact that, despite our interventions, there are currently 3 other strains of COVID19 spreading, all of which are speculated to be both more contagious _and_ more deadly. [5] While the current vaccines are _effective enough_ against the strains, there's nothing saying a new variant couldn't emerge that is resistant to the current vaccines, starting a new pandemic. And you're also ignoring the long-term cases of COVID19 [6] as well as the long term effects post-infection outside of death [7], including chronic cardiovascular damage, kidney damage, neurological effects, among many others.
So yes, the mortality rates are similar, but literally nothing else is. We don't know enough about this virus to let it run rampant, and the world should have done more from the get-go to keep it contained.
[1]: https://www.cdc.gov/flu/about/burden/past-seasons.html
[2]: https://covid.cdc.gov/covid-data-tracker/#datatracker-home
[3]: https://www.vdh.virginia.gov/coronavirus/2020/12/07/covid-19...
[4]: https://wwwnc.cdc.gov/eid/article/26/7/20-0282_article
[5]: https://en.wikipedia.org/wiki/Variants_of_SARS-CoV-2
[6]: https://en.wikipedia.org/wiki/Long_COVID
[7]: https://en.wikipedia.org/wiki/Coronavirus_disease_2019#Longe...
People with bad cases of Covid routinely spend weeks in ICU before they either get better or die. This is an enormous workload for doctors and nurses.
Also, we do not know for sure how long haulers will fare. I myself had heart arrythmia after an otherwise very mild case of Covid. Fortunately, a simple beta blocker helped me and there are no traces of anything worse on my heart, but the doctors that took care of me told me that they have a surge of youngish patients with myocarditis and similar problems. Some of those are actually sportsmen.
Annoyingly, most of what we know about long haulers is anecdotal, and long-term effects of covid are in many ways more significant than deaths.
Case Fatality Rate, in case anyone else lost their acronyms.
According to worldometers.info, the US has reported 27,101,847 cases of COVID19 and 460,398 deaths, giving a CFR of about 1.70%, so that's about right.
I have reservations about your influenza mortality number of 0.8%. Most of the numbers I've seen have been significantly lower. If you look at [1], there were an estimated 29,000,000 symptomatic cases of influenza in the US in 2016 and an estimated 38,000 deaths, giving an estimated CFR of 0.1%, including any effects of vaccination. Correcting for vaccination and making the assumption that everyone were vaccinated, that would result in a unvaccinated CFR of around 0.2-0.3, right?
"If you disagree, please be so kind and educate me why."
We've been dealing with COVID19 for about a year, no? We've had 27,000,000 cases and 460,000 deaths, right? That's about the same number of cases and 10x the number of deaths.
You may not be trying to downplay COVID19, but I suspect you are doing it.
[1] https://en.wikipedia.org/wiki/United_States_influenza_statis...
That's if we catch every case with testing. I'm not sure if this is still the case, but I've heard estimates of 25%-50%. The death toll also lags total cases 2-4 weeks.
This doesn't really relate to myths of vaccine manufacturing.
So no, we did not “have the vaccine” in February.
The author is arguing that, because it's difficult to produce of the Moderna vaccine, we couldn't have disseminated it on a wide scale in February or shortly thereafter. That argument may be fine, but it diverts attention from the critical role that the U.S. regulations in general, and the U.S. FDA in particular, played in slowing the process of widespread vaccine dissemination.
Specifically: the process could have been sped by months if the U.S. had been willing to conduct human-challenge trials [1] and to integrate tests of safety (Phase 2) and efficacy (Phase 3) [2].
To my knowledge, understanding of how to integrate the safety and efficacy trials has only recently developed; perhaps the FDA shouldn't be faulted on that front. But the benefits of human-challenge trials have long been understood.
Secondly, it would be extremely unethical to overlap phase 2 & 3 trials, when there are effective NPIs.
The lessons of unethical medical science, like Tuskegee and experimentation on armed forces personnel, must not be ignored. And yet that is precisely what your ref Eyal[1] is proposing, because practically none of the safeguards can be achieved.
Efficacy trials require a much larger participant group, given the still low infection rates. I don't see how one can justify giving a candidate to 100k people before safety screening, especially when the downside risk is significant risk of death if they ever catch COVID.
Yes, I’m not disputing this point. My claim is that even after we account for the difficulties of mass production, the FDA-controlled licensing process should have taken less time than it did.
> Efficacy trials require a much larger participant group, given the still low infection rates.
This is the major benefit of human-challenge trials [1]: if you intentionally infect healthy, well-informed, low-risk volunteers with low doses of a virus, the size of the requisite participant group falls dramatically. Note also that “Any volunteers in whom infection was confirmed would receive excellent care for COVID-19, including priority for any scarce life-saving resources, in state-of-the-art facilities.”
You can say that it’s unethical to run such trials. But it seems hard to make that argument if the volunteers are informed about the risks and if safeguards are in place (e.g., limiting the dose to which volunteers are exposed). Consider also the cost -- many thousands of lives -- of refusing to run such trials and thus delaying the licensing of a vaccine by months.
> The lessons of unethical medical science, like Tuskegee [...] must not be ignored. And yet that is precisely what your ref Eyal[1] is proposing
No, not at all. In the Tuskegee experiment, people infected with syphilis were promised medical care that they were later denied. When they were diagnosed with syphilis, they weren't even told of the diagnosis. (One of the purposes of the experiment was to study what happens when syphilis isn't treated.) These conditions are a world apart from what Eyal et al. are proposing in [1]. You don't need to take my word for it: the article is short and easy to read.
[1] https://academic.oup.com/jid/article/221/11/1752/5814216
The NPIs have enormous costs in both dollars and quality of life, and are in fact not terribly effective.
After all, all the technology is there, it's just the investment and those pesky actual manufacturing process that's standing in the way.
By analogy, it's one thing to have a fab, a working fab can make 5nm chips as designed at some yield, however, the fab is not responsible for whether the chip designs work or are valid.
My point here really was that tech people, and especially people in the software industry, ignore that the bulk of the work isn't the equipment, it's the long tale of supply chain, manufacturing process and QA. All of the 5nm equipment exists today, it's the other parts that are the hold up.
Essentially, in our own domain we know better, and we should really understand this and stop making this mistake in other domains.